bidi2pdf 0.1.14 → 0.1.16

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 57c77db5483e79f4cfb2314b792fa10eecf910ee319319c8e838cce0c14dc390
-  data.tar.gz: '029c81071a16b9c8b276afda6e162811552c095dac134d3f564497f1847840a0'
+  metadata.gz: 3533d126bc13406c70b6931112c07952aac0c19b718f3100258fb2182da8fcea
+  data.tar.gz: a761495802d5507fba3469405a571555547ed58891b216575f947e2b6a0c63be
 SHA512:
-  metadata.gz: 2eabb11610a39980898f8f66109cc61561e47058815283d7f3db711a63ddf7e88b6bb57829dc2e39cb4f0c29811086986ff27b09c3655c8b4c1281dbfa91e1c8
-  data.tar.gz: e7d8d0099c195d503593cd45fb7a62d1f94b67fd05bba6ef64181e6d239fed23fa0a9f5f934430e1b822389ce30b294f8805f0947f85dbbdfced6e4c71a5d95b
+  metadata.gz: cb29c3bf90a6a6dbd6fcf488980c51b9f666206ea18748d0c670e6fc2a64c453ca0f20043dce113094e74eb34d4b45f6be8b7d7a433522a59ac6ebab2d6ba8bb
+  data.tar.gz: bb9c384545c42d45630576b523c3c82025b1df2a2fc8975b8959276566173ad684b9423e7f30fc46b7834c61328ef68f9ddb1ee73d83b08e0ec8ea10b93f117f

data/.rubocop.yml

@@ -1,6 +1,17 @@
+# Keep RuboCop's own default excludes (vendor/**, tmp/**, node_modules/**, .git/**) when adding to
+# AllCops/Exclude below - without this, a project-level Exclude replaces them and vendor/bundle gets
+# linted.
+inherit_mode:
+  merge:
+    - Exclude
+
 AllCops:
   NewCops: enable
   TargetRubyVersion: 3.4
+  Exclude:
+    # Local Claude devbox (gitignored). Its Dockerfile.ruby ends in ".ruby", which RuboCop takes for
+    # a Ruby source file and then fails to parse.
+    - 'docker-claude/**/*'
 
 Style/StringLiterals:
   EnforcedStyle: double_quotes
@@ -79,6 +90,15 @@ RSpec/DescribeClass:
   Exclude:
     - 'spec/acceptance/**/*_spec.rb'
 
+# Matches bidi2pdf-rails' own spec/rails_helper.rb: scenario/when_ are custom
+# alias_example_group_to groups there (and here, see spec/spec_helper.rb), not the single example
+# Capybara's own `scenario` conventionally means - this cop doesn't know that and miscounts every
+# then_/and_ nested underneath as belonging to one example.
+RSpec/MultipleExpectations:
+  Enabled: true
+  Exclude:
+    - 'spec/acceptance/**/*_spec.rb'
+
 plugins:
   - rubocop-rake
   - rubocop-rspec

data/CHANGELOG.md

@@ -8,10 +8,75 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
-[unreleased]: https://github.com/dieter-medium/bidi2pdf/compare/v0.1.14..HEAD
+[unreleased]: https://github.com/dieter-medium/bidi2pdf/compare/v0.1.16..HEAD
 
 <!-- generated by git-cliff end -->
 
+## [0.1.16] - 2026-09-22
+
+### 🐛 Fixed
+- Give NavigationDNSError its own initializer
+- Enforce the complete recipe shape in --validate, not just semantic checks
+- Close schema/validator mismatches in recipe schema
+- Document and address blank PDFs from Chrome's network checks
+- Add browser tab cookie integration specs
+
+### 🔄 Changed
+- Merge pull request #134 from dieter-medium/feat/llm-friendly-cli
+- Merge pull request #133 from dieter-medium/feat/llm-friendly-cli
+- Merge pull request #131 from dieter-medium/docs/local-network-access
+- Merge pull request #130 from dieter-medium/fix/host-only-cookie-domain
+- Merge pull request #126 from dieter-medium/chore-fix-release-creds-check
+
+### 🚀 Added
+- Agent-friendly CLI - structured JSON, diagnose, and recipes
+
+## [0.1.15] - 2026-09-20
+
+### 🎨 Refactored
+
+- Remove deprecated benchmarking script
+
+### 🐛 Fixed
+
+- Handle partial Chrome session builds and cleanup
+- Ensure consistent use of alias in nginx_url helper
+- Use AtomicFixnum for thread-safe slot counting
+- Refactor Chrome session setup in specs
+- Improve session warmer configuration and timeout handling
+- Decouple chromedriver log level from logger level
+- Remove navigation ID generation for sub-resources
+- Remove redundant timeout error test
+- Improve string truncation for log safety
+- Improve release credentials check for missing keys
+- Enhance WebSocket handling for control frames and errors
+- Add chromedriver log level configuration
+- Adjust benchmark thresholds for CI environments
+- Add string truncation for safe log output
+- Set timeout for RSpec test runs
+- Prevent logging raw data URLs during navigation
+- Generate unique navigation IDs for sub-resource requests
+
+### 🔄 Changed
+
+- Merge pull request #124 from dieter-medium/feat/session-warmer
+- Merge pull request #123 from dieter-medium/feat/faster-pdf-generation
+- Merge pull request #122 from dieter-medium/chore-prepare-dev
+- Bump version to 0.1.15.pre
+
+### 🚀 Added
+
+- Implement session warmer idle time management
+- Improve session warmer threading and recovery
+- Enhance session warming and WebSocket health checks
+- Add session warmer and enhance version resolver error handling
+- Add session warming and programmatic configuration
+- Add chromedriver log level mapping and tests
+- Add release credential check for rubygems.org
+- Add navigation id handling in network events
+- Replace WebSocket client library for performance
+- Implement BufferedWebSocketClient and benchmarks
+
 ## [0.1.14] - 2026-09-19
 
 ### 🐛 Fixed
@@ -407,7 +472,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### 🔄 Released
 
-- [unreleased](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.14..HEAD)
+- [unreleased](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.16..HEAD)
+- [0.1.16](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.15..v0.1.16)
+- [0.1.15](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.14..v0.1.15)
 - [0.1.14](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.13..v0.1.14)
 - [0.1.13](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.12..v0.1.13)
 - [0.1.12](https://github.com/dieter-medium/bidi2pdf/compare/v0.1.11..v0.1.12)

data/README.md

@@ -20,15 +20,17 @@ Bidi2pdf gives you **precision, flexibility, and full control**.
 3. [Why BiDi?](#why-bidi-instead-of-cdp)
 4. [Installation](#installation)
 5. [CLI Usage](#cli-usage)
-6. [Library API](#library-api)
-7. [Architecture](#architecture)
-8. [Docker](#docker)
-9. [Configuration Options](#configuration-options)
-10. [Rails Integration](#rails-integration)
-11. [Test Helpers](#test-helpers)
-12. [Development](#development)
-13. [Contributing](#contributing)
-14. [License](#license)
+6. [Agent and Automation Usage](#agent-and-automation-usage)
+7. [Library API](#library-api)
+8. [Architecture](#architecture)
+9. [Docker](#docker)
+10. [Configuration Options](#configuration-options)
+11. [Programmatic Configuration](#programmatic-configuration)
+12. [Rails Integration](#rails-integration)
+13. [Test Helpers](#test-helpers)
+14. [Development](#development)
+15. [Contributing](#contributing)
+16. [License](#license)
 
 ## ✨ Key Features
 
@@ -39,7 +41,9 @@ Bidi2pdf gives you **precision, flexibility, and full control**.
 ✅ **Docker-ready** – Plug and play with containers  
 ✅ **Modern architecture** – Built on Chrome's next-gen BiDi protocol  
 ✅ **Network logging** – Know which requests fail during rendering  
-✅ **Console log capture** – See what goes wrong inside the browser
+✅ **Console log capture** – See what goes wrong inside the browser  
+✅ **Agent-ready** – Structured JSON output, NDJSON progress streaming, a page diagnostic, and
+declarative recipes, so LLM agents and CI can drive it without parsing logs
 
 ---
 
@@ -108,6 +112,152 @@ bidi2pdf render \
 
 ---
 
+## 🤖 Agent and Automation Usage
+
+Every command below also works without `--json` (human-readable output on stdout); with it, stdout
+carries exactly one JSON document and nothing else - safe to pipe into `jq` or parse directly.
+Human-readable logs move to stderr for the duration of any `--json`/`--output -` call, never
+mixing into stdout. `--json-stream` works independently of the two: it always writes progress
+events to stderr, whether or not `--json` is also given - in human mode, that just means
+human-readable output keeps going to stdout as usual, alongside the stream. Full JSON Schema for
+every shape is built into the gem, so an agent can discover it without reading this file:
+
+```bash
+# 1. Check compatibility first - no browser launched
+bidi2pdf version --json
+
+# 2. Discover the shape of each command's own result, a manifest, an NDJSON event, or a recipe file
+bidi2pdf schema render
+bidi2pdf schema diagnose
+bidi2pdf schema run
+bidi2pdf schema manifest
+bidi2pdf schema event
+bidi2pdf schema recipe
+
+# 3. Validate a recipe - no browser launched, the cheapest way to iterate on one
+bidi2pdf run recipe.yml --validate
+
+# 4. Run it for real
+bidi2pdf run recipe.yml --json
+```
+
+### Structured render output
+
+```bash
+bidi2pdf render --url https://example.com/invoice/14432423 --output example.pdf --json
+```
+
+```json
+{
+  "schema_version": 1,
+  "ok": true,
+  "command": "render",
+  "output": "example.pdf",
+  "bytes": 182734,
+  "sha256": "abcd...",
+  "pages": 2,
+  "duration_ms": 842,
+  "navigation": { "requested_url": "https://example.com/invoice/14432423", "final_url": "https://example.com/invoice/14432423", "status": 200 },
+  "console": [],
+  "network_failures": [],
+  "warnings": [],
+  "error": null
+}
+```
+
+`pages` is `null`, with a warning explaining why, when the optional `pdf-reader` gem isn't
+installed - see [Docker](#docker) for why the published images always have it. A failed render
+still emits exactly this shape, with `ok: false` and a structured `error` (`code`, `message`,
+`retryable`, `hint`, `details`) instead of a stack trace:
+
+| Exit code | Meaning |
+|---|---|
+| `0` | success |
+| `2` | CLI, configuration, or recipe-validation error |
+| `3` | browser or navigation error |
+| `4` | page not as expected (a recipe action/assertion, or a diagnose selector, failed) |
+| `6` | output or PDF generation failure |
+| `70` | unexpected internal error |
+
+### stdin/stdout, manifests, and progress streaming
+
+```bash
+# Render HTML piped on stdin, PDF bytes piped out on stdout - no files touched
+cat page.html | bidi2pdf render --stdin --output - > page.pdf
+
+# A render manifest: enough to reproduce and diagnose the render later
+bidi2pdf render --url https://example.com --output example.pdf --manifest render.json
+
+# One JSON progress event per stderr line as the render happens - works with human-readable
+# output too, not just --json
+bidi2pdf render --url https://example.com --output example.pdf --json-stream
+```
+
+### Diagnosing a page before trusting its PDF
+
+`bidi2pdf diagnose` loads a page like `render` does but produces no PDF - it answers *why does the
+PDF not look like the page* (console errors, failed requests, font-loading status, `@media
+print`/`@page` rules, fixed/sticky elements, Paged.js detection), not what the page's content is:
+
+```bash
+bidi2pdf diagnose --url https://example.com/invoice/14432423 --json
+```
+
+### Declarative recipes
+
+A recipe is a rendering contract - the waits a page needs before it's printed, and the properties
+the resulting PDF must have - not a general browser-automation script:
+
+```yaml
+# invoice.yml
+version: 1
+
+source:
+  url: https://example.com/invoice/123
+
+actions:
+  - wait_for:
+      selector: "#invoice"
+      timeout: 10
+  - click:
+      selector: "#show-details"
+  - wait_network_idle:
+      timeout: 10
+
+assert:
+  - selector_exists:
+      selector: "#total"
+  - no_console_errors: true
+  - page_count: 2
+  - pdf_text_present:
+      text: "Invoice #123"
+
+output:
+  pdf: invoice.pdf
+  manifest: invoice.json
+```
+
+```bash
+bidi2pdf run invoice.yml --validate   # schema + known actions/assertions, no browser
+bidi2pdf run invoice.yml --json       # actions, then the PDF, then assertions against it
+```
+
+Actions (`wait_for`, `click`, `evaluate`, `inject_script`, `inject_style`, `set_viewport`,
+`wait_network_idle`) and page assertions (`selector_exists`, `text_present`, `no_console_errors`,
+`no_network_failures`, `fonts_loaded`) are a thin layer over the same `BrowserTab` methods the
+[Programmatic API](#-programmatic-api) below uses directly. PDF assertions (`page_count`,
+`pdf_text_present`, `pdf_not_blank`) need the `pdf-reader` gem - a recipe using one fails
+`--validate` immediately, before any browser launches, when it isn't installed.
+
+`no_console_errors`, `no_network_failures`, `fonts_loaded`, and `pdf_not_blank` are presence-only
+assertions - the step is either there or it isn't, nothing reads the value beside it - so they must
+be written as `true` exactly, e.g. `- no_console_errors: true`; `false` (or any other value) is
+rejected by both `bidi2pdf schema recipe` and `--validate` rather than being silently ignored.
+`wait_for` needs exactly one of `selector`, `paged_js`, `script` - zero or more than one is
+rejected the same way, before any browser launches.
+
+---
+
 ## 🧠 Programmatic API
 
 ### Classic Approach
@@ -285,6 +435,11 @@ docker run -it --rm \
 
 ✅ Tip: Mount your local directory (e.g. ./output) to /reports in the container to easily access the generated PDFs.
 
+✅ Both published images also install [`pdf-reader`](https://github.com/yob/pdf-reader) - not a
+runtime dependency of the gem itself (see [Agent and Automation Usage](#agent-and-automation-usage)) -
+so `pages` and the `page_count`/`pdf_text_present`/`pdf_not_blank` recipe assertions work out of
+the box in either image, no extra install step needed.
+
 ### Docker Compose
 
 ```bash
@@ -331,6 +486,99 @@ docker compose -f docker/docker-compose.yml down
 | `--log_level`          | Log level: debug, info, warn, error, fatal |
 | `--remote_browser_url` | Connect to remote Chrome session           |
 | `--default_timeout`    | Operation timeout (default: 60s)           |
+| `--json`                | Emit a single structured JSON result document (see `bidi2pdf schema render`) |
+| `--json_stream`         | Emit one JSON progress event per line to stderr (see `bidi2pdf schema event`) - works with or without `--json` |
+| `--stdin`               | Read the HTML document from stdin, instead of `--url`/`--html-file`         |
+| `--output -`            | Write raw PDF bytes to stdout instead of a file                             |
+| `--manifest FILE`       | Write a render manifest (see `bidi2pdf schema manifest`) to `FILE`          |
+
+See [Agent and Automation Usage](#agent-and-automation-usage) above for `bidi2pdf diagnose`,
+`bidi2pdf run recipe.yml`, `bidi2pdf schema <kind>`, and `bidi2pdf version --json` - a separate
+set of commands with their own options, not additional flags on `render`.
+
+---
+
+## 🔧 Programmatic Configuration
+
+Beyond the per-render CLI flags above, a few gem-wide defaults are set once via `Bidi2pdf.configure`:
+
+```ruby
+Bidi2pdf.configure do |config|
+  config.default_timeout = 60 # seconds - default BiDi command timeout
+  config.enable_default_logging_subscriber = true
+  config.log_truncate_limit = 200 # bytes - see below
+  config.chromedriver_log_level = "WARNING" # see below
+end
+```
+
+| Setting                             | Default | Description                                                                                                                                                                                                                      |
+|-------------------------------------|---------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| `default_timeout`                   | `60`    | Default timeout (seconds) for BiDi commands that don't specify their own.                                                                                                                                                        |
+| `enable_default_logging_subscriber` | `true`  | Subscribes a default logger to the gem's internal instrumentation events.                                                                                                                                                        |
+| `log_truncate_limit`                | `200`   | Max bytes kept when logging a value that can be large (e.g. a `data:` URL) - truncated at a byte, not character, boundary.                                                                                                       |
+| `chromedriver_log_level`            | `nil`   | ChromeDriver's own `--log-level` (`"ALL"`/`"INFO"`/`"WARNING"`/`"SEVERE"`). Unset mirrors `Bidi2pdf.logger.level`; set explicitly to quiet ChromeDriver's own (often very verbose) output independently of your app's log level. |
+
+`Bidi2pdf.logger`, `Bidi2pdf.network_events_logger`, `Bidi2pdf.browser_console_logger`, and
+`Bidi2pdf.notification_service` are also configurable in the same block, for more advanced
+logging/instrumentation needs.
+
+### Pre-warmed sessions (`Bidi2pdf::SessionWarmer`)
+
+Optional. Keeps a few Chrome sessions started in the background so a render skips browser startup.
+Every slot is still used for exactly one render and then discarded - isolation is the same as
+launching a fresh Chrome per PDF.
+
+```ruby
+Bidi2pdf::SessionWarmer.configure do |c|
+  # warms c.size sessions right here, e.g. at boot
+  c.size = 2
+  c.max_idle_age = 300
+  # c.remote_browser_url = "http://remote-chrome:3000/session"
+end
+
+Bidi2pdf::SessionWarmer.with_tab do |tab|
+  tab.navigate_to(url)
+  tab.print("invoice.pdf")
+end
+
+Bidi2pdf::SessionWarmer.shutdown
+```
+
+| Setting              | Default               | Description                                                                                               |
+|----------------------|-----------------------|-----------------------------------------------------------------------------------------------------------|
+| `size`               | `1`                   | Number of sessions kept warm. With none ready, a render starts its own session as usual - it never waits. |
+| `max_idle_age`       | `300`                 | Seconds a warm session may sit unused before it is retired and replaced. `nil` disables the limit.        |
+| `headless`           | `true`                | Run Chrome headless.                                                                                      |
+| `chrome_args`        | `DEFAULT_CHROME_ARGS` | Chrome launch arguments.                                                                                  |
+| `remote_browser_url` | `nil`                 | Connect each slot to a remote chromedriver instead of starting a local one.                               |
+
+> **Security note:** an idle warm session is an open, unauthenticated automation endpoint
+> (chromedriver's port, Chrome's debugging port - loopback only for a local chromedriver) for as
+> long as it waits. `max_idle_age` bounds that window; keep it set unless the process runs somewhere
+> nothing else can reach those ports. With `remote_browser_url`, who can reach that endpoint on the
+> network is what matters, exactly as it does without the warmer.
+
+### Customizing Chrome arguments (and blank PDFs from inline HTML)
+
+`Bidi2pdf::Bidi::Session::DEFAULT_CHROME_ARGS` already contains one `--disable-features=...` entry,
+and Chrome only honors the **last** occurrence of that switch. To turn another feature off, extend
+that entry instead of appending a second switch, which would silently drop the defaults:
+
+```ruby
+chrome_args = Bidi2pdf::Bidi::Session::DEFAULT_CHROME_ARGS.map do |arg|
+  arg.start_with?("--disable-features=") ? "#{arg},LocalNetworkAccessChecks" : arg
+end
+```
+
+`LocalNetworkAccessChecks` is the one you are most likely to need. HTML rendered through
+`BrowserTab#render_html_content` is loaded as a `data:` URL, which Chrome treats as a public origin;
+if that HTML references assets on a private address (`localhost`, a Docker hostname, ...), recent
+Chrome versions (confirmed with Chrome 153) block those requests before they are sent. Nothing
+raises - the assets show up as failed network events, the server never sees a request, and the PDF
+comes out unstyled or blank. Disable the check only where the asset host really is private, and keep
+it on when rendering pages you do not control.
+The [bidi2pdf-rails README](https://github.com/dieter-medium/bidi2pdf-rails#readme) has the full
+walkthrough under "Blank PDFs: Chrome's Local Network Access Check".
 
 ---
 

data/docker/Dockerfile

@@ -34,7 +34,11 @@ WORKDIR /app
 # Copy your gem into container
 COPY ./pkg/bidi2pdf-*.gem ./
 
+# pdf-reader is deliberately not a bidi2pdf runtime dependency - it is installed here, as its own
+# step, so the published image always has `pages`/PDF assertions available, the environment an
+# agent or CI job actually renders in.
 RUN gem install ./bidi2pdf-*.gem && \
+    gem install pdf-reader -v "~> 2.14" && \
     chown -R appuser:appuser /app
 
 # Switch to non-root user

data/docker/Dockerfile.slim

@@ -26,7 +26,11 @@ WORKDIR /app
 # Copy your gem into container
 COPY ./pkg/bidi2pdf-*.gem ./
 
-RUN gem install ./bidi2pdf-*.gem
+# pdf-reader is deliberately not a bidi2pdf runtime dependency - it is installed here, as its own
+# step, so the published image always has `pages`/PDF assertions available, the environment an
+# agent or CI job actually renders in.
+RUN gem install ./bidi2pdf-*.gem && \
+    gem install pdf-reader -v "~> 2.14"
 
 
 # Stage 2

data/lib/bidi2pdf/bidi/browser_tab.rb

@@ -533,11 +533,21 @@ module Bidi2pdf
 
         cmd = Bidi2pdf::Bidi::Commands::BrowsingContextNavigate.new url: url, context: browsing_context_id, wait: wait
 
-        client.send_cmd_and_wait(cmd) do |response|
-          Bidi2pdf.logger.debug "Navigated to page url: #{url} response: #{response}"
+        navigation_id = client.send_cmd_and_wait(cmd) do |response|
+          logged_url = url.start_with?("data:") ? "data:[#{url.bytesize} bytes]" : url
+          response_navigation_id = response.dig("result", "navigation")
+          Bidi2pdf.logger.debug "Navigated to page url: #{logged_url} navigation: #{response_navigation_id}"
+
+          response_navigation_id
         end
+
+        check_navigation_http_status(url, navigation_id)
       rescue Bidi2pdf::CmdError => e
-        msg = e.response["message"]
+        raise_navigation_error_for(url, e)
+      end
+
+      def raise_navigation_error_for(url, error)
+        msg = error.response["message"]
         case msg
         when /^net::ERR_INVALID_AUTH_CREDENTIALS/
           raise NavigationAuthError.new(url, msg)
@@ -546,10 +556,41 @@ module Bidi2pdf
         when /^net::/
           raise NavigationError, "Connection error: #{url} #{msg}"
         else
-          raise e
+          raise error
         end
       end
 
+      # browsingContext.navigate's own response never carries an HTTP status - only a
+      # network.responseCompleted event does, correlated back to this specific navigation via its
+      # "navigation" field (a Chrome/redirect-chain-wide ID, not the network request's own id).
+      # register_event_listeners already ran above, so network_events has been tracking since
+      # before the navigate command was even sent. Deliberately conservative: with no navigation
+      # ID, or no correlated *completed* request found, this stays silent rather than guessing -
+      # only a positively confirmed >= 400 status raises. max_by(&:start_timestamp) picks the last
+      # hop of a redirect chain (every hop shares the same navigation ID), matching the page the
+      # browser actually ended up on.
+      def check_navigation_http_status(url, navigation_id)
+        return unless navigation_id
+
+        final_response = correlated_navigation_response(navigation_id)
+        return unless final_response
+
+        status = final_response.http_status_code
+        return if status < 400
+
+        raise NavigationNotFoundError, "Navigation to #{url} failed: HTTP 404 Not Found" if status == 404
+
+        raise NavigationError, "Navigation to #{url} failed: HTTP #{status}"
+      end
+
+      # The last hop of a redirect chain (every hop shares the same navigation ID) - the page the
+      # browser actually ended up on, not wherever the chain started.
+      def correlated_navigation_response(navigation_id)
+        network_events.all_events
+                      .select { |event| event.navigation == navigation_id && event.http_status_code }
+                      .max_by(&:start_timestamp)
+      end
+
       def register_event_listeners
         return if @event_handlers_registered